MEG X870E Godlike Cannot Restart via Windows or Case Button But Shutsdown Fine

Captainlooney

Member
Joined
Jan 4, 2025
Messages
31
I can shut down the system fine through Windows "Shutdown" and can turn on the PC again via the case power button just fine. The issue is when I use either the Restart button on the case or select Restart from Windows. The monitor will go blank/black but all the light are still on, fans are all still spinning, RAM RGB still on, the Dynamic Dashboard retains whatever image it had as normal and it hangs until the Dynamic Dashboard then shows "CPU & DDR Error." Outside of this, the Dynamic Display works as intended with no load issues or blank/black on startup or booting into Windows or through MSI Center. At this point I have to hard shutdown via the case power button and it will boot just fine again.

I'm ruling out the memory because extensive memory stress testing using Karhu, Memtest86, Memtest86+, OCCT all pass several hours. I've tried different configurations as well and they all exhibit the same behavior. 2x24GB, 4x24GB, 2x48GB they all pass all listed memory stress tests fine even with XMP/EXPO and manual timings. I also tried an old 2x32GB G.Skill 6400 kit but it won't memory train. For CPU, it's a 9800X3D, everything works fine whether it's gaming or running Cinebench for 30 minutes, under 72C with 26C average coolant temperature. Zero system instability observed at all even with hours of gaming. It is purely a restart issue. It does not matter, PBO enabled or default, also XMP/EXPO or default JEDEC speeds.

System is currently built in a custom water loop though the same behavior was observed when I ran with an AIO for general testing of the components it would reboot successfully less than half the time (both custom and AIO).

Current System Config:
CPU: AMD Ryzen 7 9800X3D
Motherboard: MEG X870E Godlike
RAM: 4x24GB G.Skill Trident Z5 Royal 8000 (clocked at 6400), also tried 2x48GB G.Skill Trident Z5 Royal 6400 same behavior
GPU: Asus ROG Strix 3080 Ti (for testing while waiting for a 5090 to water block)
M2_1 (boot, Windows 11): Crucial T705 4TB
M2_3 (storage): Western Digital SN850X 8TB
M2_5 (storage): Western Digital SN850X 8TB
Case: Lian-Li V3000 Plus
PSU: Seasonic PRIME TX-1600 Noctua Edition
Dynamic Display Firmware: Up to date with the latest Firmware Utility from 1/16/25
BIOS: Up to date with the latest 1/22/25 7E48v1A24

M2 slots were chosen based on slots that don't share or sap the PCI-e lanes of other components. I am using the EZ Fan Hub via the EZ Bridge cable but am also using an NZXT USB Hub in addition to another powered fan hub to support a full slew of Lian-Li LCD/LED TL Wireless fans. Dual D5 Pumps (Singularity Computers Distros) are connected to the motherboard Pump PWM headers directly.

I am not facing any weird Dynamic Display issue that folks on another thread are battling. I did set a MP4 video for the Dynamic Display so maybe I just escaped the woes that way.
 
I know asus boards have a update and a work around made right into the bios that fixed everything. Think it was called memory context restore (for asus only). We can hope msi does something similar.
I really hope we get something like this. It's way more likely to fix the issue than expecting every software to make a special change just for MSI.
 
Ok bros, just an update on the situation, I believe it is fixed, see if this can help any bros out there encountering this issue.

My spec:
Chassis: Lian Li Vision Compact (Black)
CPU: AMD Ryzen 9 9800X3D
CPU Cooler: Tryx Panorama 360 SE AIO
Motherboard: MSI MEG X870E Godlike
Memory: G.SKILL Trident Z Royal Neo 64GB (2×32GB) DDR5 6000MHz CL26 (F5-6000J2636G32GX2-TR5NG)
Storage 1: Samsung 990 Pro 4TB (PCIe Gen4)
Storage 2: Samsung 990 Pro 4TB (PCIe Gen4)
HDD 1: WD 8TB
HDD 2: WD 8TB
GPU: MSI GeForce RTX 5090 Gaming Trio OC
PSU: Seasonic Prime TX-1600 (ATX 3.0)
Fans: Lian Li TL120 Wireless
• 6× Reverse-Blade (Intake)
• 2× Normal (Exhaust)
Aesthetics & Accessories:
• Cables: Lian Li Wireless Strimer RGB (24-pin & 12V-2x6)
• Display: Turing 8.8” IPS Display
Thermal Accessories:
• Thermalright AM5 Secure Frame
• Honeywell PTM7950 40mm×40mm
—————————————————————

all software is running the latest as of 14-Apr-2025

1) I had disconnected all usb devices except for mouse and keyboard (including rear I/O, and usb headers all unplugged)

2) closed all monitoring softwares, uninstalled L-Connect 3 version 2.0.24 (deleted L-Connect 3 folder after uninstalled), disabled all monitoring software auto start

3) used MSI center to scan live update, updated chipset, updated Realtek HD audio, updated LED thingy, updated power something, updated Vbios

4) updated windows

5) reboot failed like it normally does, hard shut down with power button, cold boot back to finish windows updates, back to windows clean (no monitoring softwares at all)

6) shut down, cleared cmos, turned off PSU power, unplugged PSU, waited 5 mins

7) boot into BIOS, back to default profile, no OC, no PBO, no EXPO, boot into windows, set up all Microsoft visual studio for L-Connect 2.0.26 (I didn’t had 2010 x64 & x86 installed, only had 2012 & 2015-2022), after confirmed I have all visual studio installed, then shut down

8) flashed MSI bios A30

9) set up as normal

10) back into windows, plugging back usb devices one by one, with the related monitoring software one by one, each usb device with the software enabled, I had tested each with windows restarted NO ISSUE!!! (DO NOT PLUG THEM BACK ALL AT ONCE, do it one by one with software related one by one, test to ruled out if any one of those is causing issue)

11) shut down, plugging back all usb devices, plugging back usb header devices

12) cold boot, back to windows with monitoring app enabled, ran a few stress tests, updated DD3 data (only data), restart, NO ISSUE, IT FXXKING RESTARTED WITH NO ISSUE!!!!

13) restarted 20 something times, zero issues!!!!!!!!!!!

HOLY FXXK!!! the PC is normal now!!!! for some reason I am missing the days that I had to cold boot, missed pressing the button 😂

Monitoring softwares I have installed and enabled auto start:
MSI Center (mystic light, DD3)
MSI Afterburners latest beta
L-Connect 3
Tryx Kanali
 
Last edited:
My question is why did that fix it? It almost seems like it's not a fix but you did something in all of those steps to avoid a race condition
the board was on bios A20 out of the box, updated to A24 no fix, avoided A27, A30 was recently out, I had set up the system like anyone would had done it out of the box, things happens, as well as the other thread related to DD3, this thread is focused on restart 0D, what ever fixed that issue, I had no idea, that’s all the steps I did
 
Just to close the loop on this thread, it was indeed Lian-Li's L-Connect 3 software that was preventing the shutdown. At least as of 2.0.28 of L-Connect 3, they have swapped hardware monitoring SDK to HWINFO64 and it no longer prevents reboots via Windows or the case's reset button. The only downside is it clashes with HWINFO64 if you have them start at the same time.
 
Just to close the loop on this thread, it was indeed Lian-Li's L-Connect 3 software that was preventing the shutdown. At least as of 2.0.28 of L-Connect 3, they have swapped hardware monitoring SDK to HWINFO64 and it no longer prevents reboots via Windows or the case's reset button. The only downside is it clashes with HWINFO64 if you have them start at the same time.
I'm using a few different softwares including SignalRGB (which is a must for me) and ASUS Armory Crate (another must for my AIO screen).

This issue is not fixed for me yet.
 
Only thing that solve this issue for me was in HWinfo64 turn off SMbus support and EC support on settings its seems to be an issue caused by some monitor programs that put data on Embedded Controller that dont clear with reboot couse of bios issue
 
Back
Top